Like ﬁlings to a magnet, issues of all shapes, sizes, and degrees of importance are sticking to the idea of neuroethics. Martha Farah, an early thinker in this new ﬁeld, proposes that, numerous as they are, the problems actually fall into just three categories. She ﬁnds that neuroethics has made a quick start sizing up many practical—and some unique—questions swirling up from brain science, but, she writes, watch for challenges that reach beyond these to the metaphysical. Neuroscience may one day explain in terms of neural tissue virtually all aspects of human cognition and emotion—realms traditionally deemed apart from physical law. Thus, we should also expect neuroethics to grapple with our fundamental distinction between persons and mere “things.” If mental processes prove to result from purely physical events, this opens to question our notions of consciousness, spirituality, free will, and moral responsibility.
Despite its compact name, neuroethics encompasses a large and varied set of issues. In its tremendous diversity, can we ﬁnd a framework for thinking about the relations among issues in neuroethics? To me, the perplexing array of questions actually sorts usefully into categories according to whether they are examples of already familiar bioethical problems or arise uniquely in connection with the brain, and whether the problems are of mainly practical relevance or are primarily philosophical.
BIOETHICS AND NEUROETHICS
The traditional bioethical concerns of informed consent, privacy, the weighing of beneﬁts, and other problems in medical science all have their neuroscience counterparts. In many cases, neuroethics can directly apply bioethical analyses developed in other areas of biomedical research. These are the classic bioethical issues of neuroethics, problems for which the ethical analyses are substantially the same, whether the organ system in question is liver, kidney, or brain. For example, as the sensitivity of neuroimaging improves, researchers must prepare to deal with the occasional dilemma of disclosing an incidental ﬁnding that a research volunteer has a sign of disease or injury that wasn’t the object of the study. Such ethical quandaries are not unique to studies of brain function but concern privacy and consent in ways familiar within biomedical research in general. Recognizing which neuroethical problems are new manifestations of classical bioethical problems, and bringing the appropriate precedents to bear, is an important task on which neuroscientists and bioethicists must collaborate.
Other neuroethical issues differ substantially from those in bioethics. What makes neuroethics distinctive is the unique relation between brain function and the mind. Neuroimaging opens a window into the psychology of individuals that is unprecedented in genetic testing or any other biomedical procedure. Although privacy is a central issue in classical bioethics, protection of mental privacy— the freedom to think one’s own thoughts unobserved—adds a signiﬁcant new dimension to the discussion. Similarly, the ability to manipulate cognitive and affective states neurochemically raises a host of questions beyond the classical bioethical debate about therapeutic interventions versus enhancement. The manipulation of an individual’s psychology, and the behaviors that ensue from it, has unique social and ethical implications not shared by the manipulation of other aspects of a person’s biology.
THE PRACTICAL VS. THE METAPHYSICAL
A commonality among the foregoing issues is that they all have practical consequences. How we analyze these concerns and develop policies to address them will affect our lives in tangible ways. Much of the nascent neuroethics literature is devoted to such practical questions: anticipating the uses to which neuroscience can be put and analyzing their consequences for the individual and society.
In contrast, provocative metaphysical questions arise from the ways in which neuroscience challenges our understanding of human behavior. Of course, understanding behavior inﬂuences our policies and actions too, but such consequences are secondary to a more fundamental reordering of our concepts of mental and physical. A major task in neuroethics is to reconcile our advancing knowledge of mind-brain relations with the traditional metaphysics of dualism, that is, the idea that brains are material systems whereas minds are something very different.
Simple intuition, as well as the work of many great philosophers, points to a basic metaphysical distinction between persons—who have minds, are conscious, and behave for reasons—and the physical matter of the body, which behaves according to the laws of physics. However, this distinction becomes harder to maintain in the face of the advancing neuroscience of cognition and emotion, which seems to leave no aspect of human psychology outside the realm of physical systems implemented in neural tissue.
BRAIN IMAGING: A NEW PRIVACY FRONTIER
While functional brain imaging presents classical bioethical issues (foremost among them risk and consent for procedures involving radiation or high magnetic ﬁeld strengths and the possibility of discovering incidental brain anomalies), distinctive neuroethical problems of a practical nature arise from neuroimaging. These concern the potential for reading psychological states and traits from images. The uses for such psychologically sensitive information would be plentiful, and practical ethical questions abound, ranging from who might legitimately access it to the privacy rights of individuals whose psyches were under scrutiny.
“Neuromarketing,” for example, involves the use of brain imaging to measure consumers’ desire for a product. Some of the brain areas found to be responsible for the cravings experienced by drug abusers seeking a “ﬁx” can also be activated, albeit more weakly, by the sight of legal products. To the extent that neuroimaging can measure desire for a product, it provides a valuable new kind of information for marketers. In one famous study, subjects’ liking for Coke versus Pepsi was found to depend on the taste of the soda and also the brand name, and brain activation correlated with both sources of appeal.1
To the extent that neuroimaging can measure desire for a product, it provides a valuable new kind of information for marketers.
Another potentially privacy-breaching use for functional imaging of brain states is lie detection. Although lie detection based on functional magnetic resonance imaging is still far from feasible in real-world situations, researchers have found correlates of deception in the laboratory. A different technique, measuring brain waves called “event-related potentials” (ERPs), comes closer to providing actual brain-based lie detection. ERPs have been used to identify “guilty knowledge” by distinguishing responses to items that are generally known to be associated with a crime and items that only the perpetrator would know are associated.2 This method, called Brain Fingerprinting by its developer, has been admitted as evidence in court and is being promoted as a means of screening for terrorists.
Brain imaging can also provide mental trait information that is in many ways analogous to genetic information about personal characteristics and vulnerabilities. Like genotyping, “brainotyping” can reveal information about mental health vulnerabilities and predilection for violent crime. Unconscious racial attitudes are manifest in brain activation. 3 Sexual preferences can in principle be determined based on the ﬁnding that sexual attraction and even the attempt to suppress feelings of attraction have neuroimaging correlates.4 A growing body of literature has investigated the neural correlates of personality using brain imaging, including extroversion and neuroticism, risk-aversion, pessimism, persistence, and empathy.
Like genotyping, “brainotyping” can reveal information about mental health vulnerabilities and predilection for violent crime.
Of course, current functional neuroimaging cannot determine personality with any precision (nor can genotyping for that matter). Brain imaging is at best a rough measure of personality, but this is not to say it is uninformative even in its current state of development. A recent review of the literature found that scanning protocols developed for studying personality-brain correlations in small groups of subjects would nevertheless allow us to narrow the range of possible levels of a personality trait on the basis of a subject’s brain activity alone.5
Imaging may eventually provide us with sensitive and speciﬁc measures of psychological processes. For now, such uses must be approached carefully and with a healthy dose of skepticism.
Of course, to the extent that brain imaging is not ready to deliver ethically sensitive psychological information about people, we face a different ethical problem: The public tends to view brain scans as more accurate and objective than in fact they are. Statements like “the brain does not lie” crop up in popular writing on neuromarketing and brain-based lie detection. Although brain-based measures are one causal step closer to psychological states and traits than behavioral or autonomic measures and could therefore be viewed as more direct measures of these traits and states, they are far from straightforward. Many layers of signal processing, statistical analysis, and interpretation separate image data from the psychological inferences they support. Imaging may eventually provide us with sensitive and speciﬁc measures of psychological processes. For now, such uses must be approached carefully and with a healthy dose of skepticism.
BRAIN ENHANCEMENT: SAFETY, FREEDOM, AND FAIRNESS
Psychotropic drugs, including antidepressants, antianxiety drugs, and stimulants, are among the most widely taken prescription medications in the world. With improving side-effect proﬁles, increased public awareness of mental illness, and aggressive marketing of psychiatric medications to physicians and patients, these medications are now widely used by people who would not have been considered ill two decades ago. Indeed, psychopharmacology is now routinely used to enhance normal brain function as well as treat disease, raising many ethical issues, some practical and some metaphysical.
Peter Kramer’s landmark book, Listening to Prozac, was the ﬁrst to call attention to the growing practice of brain enhancement. He recounted the decisions of a number of patients to resume taking their antidepressants despite their successful recoveries from depression or other illnesses, because on the medication they felt “better than well.”
Surprisingly little scientiﬁc research has been done on the effects of these drugs on people who are not depressed. It seems clear antidepressants are not happy pills, shifting depressed people to normalcy and normal people to bliss. Rather, for most people they seem to leave positive feelings unchanged but attenuate negative feelings— for example reducing the subjectively experienced “hassle” factor of life. They also have subtle effects on social behavior. Sleep, eating, and sexual behavior can all be inﬂuenced pharmacologically, and a large demand exists for ways of enhancing these functions. The wakefulness-promoting agent modaﬁnil, approved in the United States for treatment of certain sleep disorders, is prescribed off-label for a panoply of other conditions and is said to be favored by some ambitious professionals as a way of packing more work into a day.6 A number of different appetite suppressants are under development, and if one of them proves safe and effective it will undoubtedly achieve “blockbuster” status. Finally, although sildenaﬁl (Viagra) and more recent medications for erectile dysfunction do not achieve their effects by altering brain function, newer neurally active drugs are in development, aimed at improving both male and female libido. If society’s experience with sildenaﬁl is any indication, many people without sexual dysfunction will seek these drugs to enhance their sex lives.
Although methylphenidate (Ritalin) and amphetamine (Adderall) are ostensibly prescribed mainly for the treatment of ADHD, sales ﬁgures suggest that they are not uncommonly used for enhancement.
Two main cognitive systems have been targeted for enhancement: attention and memory. Stimulant medication used for attention deﬁcit hyperactivity disorder (ADHD) also enhances normal performance on a variety of attentional measures. Although methylphenidate (Ritalin) and amphetamine (Adderall) are ostensibly prescribed mainly for the treatment of ADHD, sales ﬁgures suggest that they are not uncommonly used for enhancement. Surveys have estimated that as many as 10 percent of high school students and 20 percent of college students have used prescription stimulants such as Ritalin illegally.7
A major research effort is directed to the development of memory-boosting drugs targeting various stages of memory encoding. Although this research is aimed at ﬁnding treatments for dementia, some of the products under development are believed to have the potential to enhance normal memory as well, particularly in middle and old age when a degree of increased forgetfulness is normal. Finally, the ability to weaken or prevent the consolidation of unwanted memories constitutes another kind of enhancement under development.
Methods for altering brain function without drugs have also evolved rapidly over the past decade and in the future may offer complementary approaches to enhancement. Transcranial magnetic stimulation (TMS) has moved from lab to clinic as a means of treating depression and is being explored with healthy subjects as a means to alter mood and cognitive style. More invasive methods such as surgery, brain and vagus nerve stimulation, and brain-machine interfaces may eventually expand our conception of brain enhancement yet further.
The practical neuroethical issues abounding in brain enhancement can be grouped into two general categories. In the ﬁrst are health issues: safety, side effects, and unintended consequences. Our tolerance for risk is lower for enhancement than for therapy, and in comparison to other enhancements such as cosmetic surgery, neuroscience-based enhancement intervenes in a far more complex system. We are at greater risk of unanticipated problems when we tinker.
In the second category are social issues: How will brain enhancement affect the lives of all of us, including those who may prefer not to enhance? For example, the freedom to remain “au naturel” may be difﬁcult to maintain in a society where one’s competition is using enhancement. Conversely, barriers such as cost will prevent some who would like to enhance from doing so. This would exacerbate the disadvantages already faced by people of low socioeconomic status in education and employment.
BRAIN ENHANCEMENT: PERSONAL IDENTITY AND JUST DESERTS
Brain enhancement also highlights the metaphysical difﬁculty of reconciling personhood with brain function. Persons and things can both change, but they do so in different ways. Persons typically change as a result of learning, practice, or will power. Things cannot change by these means; instead they change by physical manipulations such as adding or rearranging parts. But improving mood or attention pharmacologically is a lot more like adding parts than learning or trying. In this sense, brain enhancement erodes the fundamental distinction between things (even complex biophysical things) and persons. The worry is that we are treating people (including ourselves) as objects if we chemically upgrade their cognition, temperament, or sexual performance.
If we fall in love with someone who is on Prozac and then ﬁnd she is difﬁcult and temperamental off the drug, do we conclude we don’t love her after all?
The metaphysical issues raised by brain enhancement are plain when we think about personal identity and just deserts. If we fall in love with someone who is on Prozac and then ﬁnd she is difﬁcult and temperamental off the drug, do we conclude we don’t love her after all? Then who was it we loved? Have we “cheated” if we study better with Ritalin, or can we take credit for our improved work?
PERSONAL RESPONSIBILITY AND THE BRAIN
Early observations of prefrontal-damaged patients who developed antisocial characteristics led to the thriving ﬁeld of social cognitive neuroscience, which includes lesion and imaging studies of decision making, self-control, and empathy. Certain prefrontal regions play a crucial role in these abilities, and damage to these regions may lead to irresponsible and selﬁsh behavior.
How do these scientiﬁc advances affect our understanding of moral and legal responsibility? We do not blame people for acts committed reﬂexively (for example, as the result of a literal knee-jerk), in states of diminished awareness or control (such as while sleepwalking or under hypnosis), or under duress (as with a gun held to the head), because in these cases we perceive the acts as not resulting from the exercise of free will. The problem with neuroscience accounts of behavior is that everything we do is like the knee-jerk reﬂex: resulting from a chain of purely physical events that are as impossible to resist as the laws of physics.
Furthermore, our intuitions about responsibility and blame are more inﬂuenced by the existence of a speciﬁc physical mechanism than by the abstract principle that some physical mechanism must be at work. For this reason, we are not inclined to blame the famous nineteenth-century brain injury victim, Phineas Gage, for his bad behavior after an inch-wide, 3-foot-long iron bar was blown through his head, damaging his ventromedial prefrontal cortex and transforming his personality from responsible and polite to slothful and ill-tempered. The challenge arises when we try to draw a principled line between the causes of bad behavior by someone like Gage and bad behavior that lacks obvious neurological causes.
Progress in neuroscience is contributing to this conundrum, too, by showing that the brain areas involved in behavior can be damaged in subtle and gradual ways in addition to the obvious damage caused by ﬂying iron bars. Most illicit drugs affect these areas, and prolonged drug abuse has been linked to impaired prefrontal function. Even childhood abuse or severe neglect, which involve neither a direct mechanical insult to the brain nor a foreign substance crossing the blood-brain barrier, damages these systems.8 Lastly, of course, genetic factors inﬂuence the function of these systems. This puts us on a slippery slope, however, once we recognize that all behavior is 100 percent determined by brain function, which is in turn determined by the interplay of genes and experience. The growing awareness of neuroscience explanations of criminal behavior has prompted ethicists and legal theorists to seek other interpretations of responsibility that do not depend on free will. For example, University of Pennsylvania law professor Stephen Morse has argued that rationality, rather than free will, is what makes people responsible for their behavior. The “disease model” of substance addiction and extension of the medicalized notion of addiction to other compulsive behaviors, such as compulsive gambling and compulsive sex, is another way in which brain-based explanations of behavior have changed the way we think about moral responsibility.
“IS THAT ALL THERE IS?” —SPIRITUALITY AND THE BRAIN
Most people believe in some essence of a person that is more than just the hundred or two hundred pounds of matter we can see and touch. Yet as neuroscience advances, the human mind is increasingly understood to be no more than the functioning of a material system. This ﬁrst became clear in the realms of perception and motor control, where mechanistic models have been under development for decades. However, such models do not seriously threaten our intuitively “dualist” view of mind and brain. You can still believe in “the ghost in the machine” and simply conclude that color vision and gait are features of the machine rather than the ghost.
However, as neuroscience begins to reveal the mechanisms of personality, this interpretation becomes strained. The brain imaging work already described indicates that important aspects of our individuality, including the psychological traits that matter most to us as people, have physical correlates in brain function. Pharmacologic inﬂuences on these traits also remind us that human personality has physical bases. If a selective serotonin reuptake inhibitor (SSRI) can help us take everyday problems in stride, and if a stimulant can help us meet our deadlines and keep our commitments at work, then mustn’t unﬂappable temperaments and conscientious characters also be features of people’s bodies? And if so, is anything about people not a feature of their bodies?
Recent neuroimaging research has shown a characteristic pattern of brain activation associated with states of religious transcendence, which is common to Buddhist meditation and Christian prayer.A dualist might answer that consciousness and spirituality are not physical. Yet neuroscience is making inroads with these mental phenomena, too. Research on consciousness in brain-damaged patients and normal individuals has succeeded in establishing reliable neural correlates of conscious awareness. Recent neuroimaging research has shown a characteristic pattern of brain activation associated with states of religious transcendence, which is common to Buddhist meditation and Christian prayer.10
Scientists and theologians have long struggled with the challenge of maintaining religious beliefs while accepting science’s view of the natural world. The idea that a person is somehow more than his or her physical instantiation runs deep in the human psyche and is a central element in virtually all the world’s religions. Neuroscience has begun to challenge this view by showing that not only perception and motor control but also character, consciousness, and sense of spirituality may be features of the machine. If they are, then why suppose the existence of a ghost in there at all?
The incompatibility between the intuitive or religious view of people and the neuroscientiﬁc view joins the list of metaphysical incompatibilities between persons as continuous in time and brains as changeable pharmacologically, between persons achieving through effort and brains operating better or worse as a function of neurotransmitter levels, and between persons acting intentionally and brains reacting according to physical law.
NEEDING TO “THINK WELL”
The topics reviewed here are just a sample of the large and growing set of ethical issues raised by progress in neuroscience. I have tried to reduce the perplexity one naturally feels when encountering such a complex new ﬁeld by organizing the issues into some general categories. Of course, the ﬁeld’s complexity is likely to defy any simple system of categories, and there is nothing hard and fast about the categories I have proposed.
The practical and metaphysical issues of neuroethics differ in terms of the immediacy and concreteness of the ethical problems raised, but even the issues classiﬁed as metaphysical have clear implications for society.
For example, just as the classical bioethical issues of neuroethics have useful precedents from other areas of biomedical research, so, too, are helpful precedents to be found for the distinctive neuroethical issues I described, although the correspondences and analogies may be less exact. Similarly, the practical and metaphysical issues of neuroethics differ in terms of the immediacy and concreteness of the ethical problems raised, but even the issues classiﬁed as metaphysical have clear implications for society. For example, our laws and mores are based on an understanding of why humans behave for good or ill, and the thoroughly materialist account of behavior offered by neuroscience will undoubtedly inﬂuence our legal systems and social norms. Similarly, the existence of an immaterial soul is a matter of religious belief that, like the origin of the human species, has the potential to divide society and incite conﬂict. Everything we can do to “think well” about all these dilemmas—classical, practical, and metaphysical—will help us reduce their divisive potential in our lives as individuals and as a society. Acknowledgments: The author’s studies in neuroethicsare supported by NIH grants R21-DA01586, R01-HD043078, R01-DA14129, and NSF grant 0342108.